跳到正文
智迹闻AuraTracer

发布于

Beam 1.0

公开成绩与计算依据

综合参考分根据已公开的评测资料整理。查看发布日期依据

综合参考分 38.4 · 11项基准

推理求解 32.9

Terminal-Bench-Science · 1.43%

Vals AI · 独立评测 · 0.1.0

Beam 1.0 / Mini-SWE-agent · Current displayed evaluator row. Mini-SWE-agent. Scores are not pooled across harnesses; source methodology and repeat counts retained.

HLE · 0.128 比例

Artificial Analysis · 独立评测 · Diamond

Beam 1.0 · Artificial Analysis current published protocol; hle-diamond

CritPt · 0.265 比例

Artificial Analysis · 独立评测 · 2 Hard

Beam 1.0 · Artificial Analysis current published protocol; critpt-v2-hard

工具与任务执行 50.3

GDPval-AA · 1177.08 分

Artificial Analysis · 独立评测 · 2.1

Beam 1.0 · Artificial Analysis current published protocol; gdpval-aa-normalized

Terminal-Bench · 0.0303030303030303 比例

Artificial Analysis · 独立评测 · 4.0

Beam 1.0 · Artificial Analysis current published protocol; terminalbench-4-0

AutomationBench · 0.4845786522065588 比例

Artificial Analysis · 独立评测 · current published task suite

Beam 1.0 · Artificial Analysis current published protocol; automationbench-aa

AA-Briefcase · 976.2 分

Artificial Analysis · 独立评测 · 1.1

Beam 1.0 · Artificial Analysis current published protocol; aa-briefcase

编程 20.1

FrontierCode · 0.0553143939393939 比例

Artificial Analysis · 独立评测 · 2.0

Beam 1.0 · Artificial Analysis current published protocol; frontiercode-2-0

Vals Code Migration · 22.85%

Vals AI · 独立评测 · 1.1

Beam 1.0 / Published Code Migration 1.1 sandbox · Current displayed evaluator row. Published Code Migration 1.1 sandbox. Scores are not pooled across harnesses; source methodology and repeat counts retained.

语言与信息处理 48.7

GDP.pdf · 0.116 比例

Artificial Analysis · 独立评测 · current published task suite

Beam 1.0 · Artificial Analysis current published protocol; gdp-pdf

AA-Omniscience · 0.30733333333333335 比例

Artificial Analysis · 独立评测 · 1.1

Beam 1.0 · Artificial Analysis current published protocol; omniscience-accuracy

AA-Omniscience · 0.5899903753609239 比例

Artificial Analysis · 独立评测 · 1.1

Beam 1.0 · Artificial Analysis current published protocol; omniscience-non-hallucination

查看原始成绩与计算数据

资料来源

模型资料来源 · 目录许可